Unknown revision esp32? revision (unknown 0xa)

[Cellie]

Today I received some new MH-ET mini kits.

With esptool 2.3.1 ( on Linux Mint 17.2 Rafaela ) I get this output when I flash the chip:

Code: Select all

Connecting.....
Chip is ESP32D0WDQ6 (revision (unknown 0xa)
Features: WiFi, BT, Dual Core, VRef calibration in efuse

The markings on the RF can are also a bit different from the ones I previously received.

I am at the latest Arduino core. v3.1-dev-661-gf586f5e6

Is there a third revision chip? I searched the net and this site, but got no results.

[Ritesh]

Today I received some new MH-ET mini kits.

With esptool 2.3.1 ( on Linux Mint 17.2 Rafaela ) I get this output when I flash the chip:

Code: Select all

Connecting.....
Chip is ESP32D0WDQ6 (revision (unknown 0xa))
Features: WiFi, BT, Dual Core, VRef calibration in efuse

The markings on the RF can are also a bit different from the ones I previously received.

I am at the latest Arduino core. v3.1-dev-661-gf586f5e6

Is there a third revision chip? I searched the net and this site, but got no results.

Hi,

I checked latest ECR document and different modules datasheet as well as ESP32 Technical Reference Manual and found only two revisions.

So, I don't think they have released 3rd revision without any information.

[Ritesh]

Hi Espressif Systems Developer,

Is it possible to get this type of modules into ESP32 based module lot?

Because we are planning to use ESP32 based module into one of our running product in which ESP8266 based module is used.

[WiFive]

If set, the ESP32's maximum CPU frequency has been rated

It's just a rev1 rated at 240mhz

[ESP_Angus]

Hi Cellie,

Thanks for reporting this and sorry for the confusion. This is a bug in esptool.py. The chips are revision 1 silicon, but only recently have had the efuse bits set to confirm the CPU frequency rating (previously these efuses were not set).

As WiFive says your chip is revision 1 silicon, 240MHz rated. Will fix the bug in esptool.py ASAP.


Angus

[Cellie]

@ESP_Angus & WiFive: Thanks for the explanation. Made things clear.

For reference (from https://github.com/espressif/esp-idf/bl ... .h#L94-L99 ):

Code: Select all

/* EFUSE_RD_CHIP_CPU_FREQ_RATED : R/W ;bitpos:[13] ;default: 1'b0 ; */
/*description: If set, the ESP32's maximum CPU frequency has been rated*/
#define EFUSE_RD_CHIP_CPU_FREQ_RATED  (BIT(13))
#define EFUSE_RD_CHIP_CPU_FREQ_RATED_M  ((EFUSE_RD_CHIP_CPU_FREQ_RATED_V)<<(EFUSE_RD_CHIP_CPU_FREQ_RATED_S))
#define EFUSE_RD_CHIP_CPU_FREQ_RATED_V  0x1
#define EFUSE_RD_CHIP_CPU_FREQ_RATED_S  13

[ESP_Angus]

Thanks for posting the link to the relevant registers for people, Cellie. Much appreciated!

This bug is fixed on esptool.py's github master branch now, and will be in the forthcoming esptool v2.4 release.

[Ritesh]

Hi,

I have checked 4 different ESP32-WROVER-I modules using esp efuse 2.3 versioned python script and getting correct values of silicon revision as 1 while executing efuse script with summary as command.

Below is snapshot of revision for efuse summary read command.

Identity fuses:
MAC MAC Address
= 30:ae:a4:76:b2:40 (CRC 30 OK) R/W
CHIP_VER_REV1 Silicon Revision 1 = 1 R/W (0x1)
CHIP_VERSION Reserved for future chip versions = 0 R/W (0x0)
CHIP_PACKAGE Chip package identifier = 0 R/W (0x0)